Doug Stewart wrote:
Doug Stewart wrote:
Richard wrote:
In article <45E10426.2010300@sympatico.ca>, Doug Stewart <dastew@sympatico.ca> writes:
I ran the sunfish par the Mr. Fisher sent out recently and I liked the colour map so I saved it and then tried to use it and found that it had changed.
None of the colormap stuff has changed, so if you could email me a series of steps that reproduces this problem, that would be great.
In Windows the colors are in the full range of 0-255 per channel, but in the DOS/VGA world you only have 0-63 per channel and there was some code there that assumed that 0-63 was the right range, so I adjusted that code so that the colors would display properly.
I then found that as soon as you start colour cycling the colours change.
When I look at the two pictures you referenced, I don't see that the colors themselves are different, they just look cycled. Color cycling just moves the entries around in the palette, it doesn't change the entries themselves.
OK? when I look at those 2 pictures they are very different. First) of all it is shifted as you say but only by about 5 spots. then look closely and you will see groups of 3 the same and then 1 different. Second) there is a red spot near the middle of the second picture and in the original there is no red all by itself.
I tried this on 2 different machines so I am sure its not just my machine.
I just found a clue . When I start color cycling with the - button then the problem shows up. but if I cycle with the + button all is well. The - direction also goes 3 times faster than the + direction. Doug
the easiest way to see this is to 1) make any fractal -- Mandel is ok 2) start cycling with the + 3) do shift F1 this is a gray map 4) press the - key 5) you should see a red and yellow show up that should not be there. 6 ) let it keep cycling - each time through the complete map you should see more red. the more red was this way on my machine. Cycle # # of reds 1 1 2 2 3 9 4 13 5 17 etc. Doug